PulseExploreJournal ClubResearchersJournals
Instagram
HomeJournal ClubExplore
Synapse
⌘+K
Synapse
January 1, 2022SHILAP Revista de lepidopterologíaOpen Access

XGBoost, A Novel Explainable AI Technique, in the Prediction of Myocardial Infarction: A UK Biobank Cohort Study

View Full Paper
Ask AI
Bookmark
Share

Why the study?

The authors sought to assess whether explainable AI using extreme gradient boosting (XGBoost) could outperform traditional logistic regression in predicting myocardial infarction in a large cohort.

Does an XGBoost machine learning model improve the prediction of myocardial infarction compared to traditional logistic regression in a large population-based cohort?

Population

502 506 volunteers aged 40 to 69 years in the UK Biobank

Comparison

XGBoost vs traditional logistic regression

Design

Population-based prospective cohort study

Follow-up

Until end of 2019

Authors

AMAlexander MooreMBMax Bell

Discussion

Loading...

Member takes

Overview

XGBoost may improve MI prediction over logistic regression; leaves open its role in routine cardiovascular risk assessment.

Structured PICO

Does an XGBoost machine learning model improve the prediction of myocardial infarction compared to traditional logistic regression in a large population-based cohort?

P
Population
502,506 volunteers with active consent, aged 40 to 69 years at recruitment from 2006 to 2010, from the UK Biobank population-based prospective cohort.
I
Intervention
Extreme gradient boosting (XGBoost) machine learning model with SHAPley value visualization for predicting risk of myocardial infarction.
C
Comparator
Traditional logistic regression model.
O
Outcome
Myocardial infarction (followed until end of 2019).hard clinical

XGBoost machine learning models provide superior predictive accuracy for myocardial infarction compared to traditional logistic regression, offering an explainable AI approach for cardiovascular risk assessment.

Cite This Study

Moore et al. (2022) studied this question.

synapsesocial.com/papers/69dcc304f3d3790cb7133a8fhttps://doi.org/10.1177/11795468221133611
View Full Paper
Ask AI
Bookmark
Share